Design and application of Type-2 fuzzy logic system based on improved ant colony algorithm

Abstract

In this paper, an improved ant colony optimization (IACO) with global pheromone update is proposed based on ant colony optimization (ACO), and it is used to design interval Type-2 TSK fuzzy logic system (FLS), including parameters adjustment and rules selection. The performance of the system can be improved by obtaining the optimal parameters and reducing the redundant rules. In order to verify the feasibility of the proposed method, the intelligent FLS is applied to predict the international petroleum price and the Zhongyuan environmental protection shares price. It is proved that the IACO can improve the efficiency of the original algorithm and accelerate the convergence speed. The simulations show that both IACO and ACO are feasible and have a high performance for the design of FLS. The simulation results compared with back-propagation design (BP algorithm) show that intelligent algorithms have an advantage over the classical algorithm, the simulation result compared with without rule-selection shows that reduced redundant rules can improve the performance, and the result compared with the Type-1 FLS shows that interval Type-2 TSK FLS has a better performance than the Type-1 TSK FLS.